molecular analysis and growth evaluation of northern pike

Aquaculture, 103 (1992) 253-273 Elsevier Science Publishers B.V., Amsterdam

253

Molecular analysis and growth evaluation of northern pike (Esox lucius) microinjected with

growth hormone genes

Mark L. Grossa**, John F. Schneider” *d**, Neomi Moavbli, Boaz Moavevf, Carmen Alvarezblg, Steve H. MysteP, Zhanjiang Liu’, Eric M. Hallermanh, Perry B.

Hackettdle, Kevin S. Guisebvd, Anthony J. Farascsd and Anne R. Kapuscinski” “Fisheries and Wildlife, bDepartments of Animal Science, ‘Microbiology, *The institute of Human

Genetics, and ‘Genetics and Cell Biology, University of Minnesota, St. Paul, MN, USA /Department of Zoology, Tel-Aviv University, Ramat-Aviv, Israel

gDepartamento de Biologia Celular y Genetica, Universidad de Malagas, Campus Universitario de Teatinos, Malaga, Spain

hDepartmen f t o Fisheries and Wildlife Sciences, Virginia Polytechnic Institute and State University, Blacksburg, VA, USA

‘Biotechnology General Kiryat Weizmann, Rehovot, Israel

(Accepted I 1 September 1991)

ABSTRACT

Gross, M.L., Schneider, J.F., Moav, N., Moav, B., Alvarez, C., Myster, S.H., Liu, Z., Hallerman, E.M., Hackett, P.B., Guise, K.S., Faras, A.J. and Kapuscinski, A.R., 1992. Molecular analysis and growth evaluation of northern pike (Esox lucius) microinjected with growth hormone genes. Aquaculture, 103: 253-273.

Bovine (bGH) or chinook salmon (Oncorhynchus tshawytscha) growth hormone (csGH) cDNA genes were transferred by microinjection into newly fertilized northern pike (&OX lucks) eggs. Non- lethal screening of fin tissue showed genomic integration of transgenes in 88 of 1398 putative trans- genie fish. Expression of bGH transgenes under transcriptional control of the Rous sarcoma virus long terminal repeat was detected in 36 of 1218 putative transgenic fish examined by radioimmunoassay of blood serum, Bovine growth hormone was also detected in mesodermal tissue of fins from microinjected fish using thin slice immunohistochemistry. Southern hybridizations of six tissues from a sample of 40 microinjected individuals revealed a high degree of mosaicism, with 30% of the fish containing detectable transgenic DNA in one or more tissues and only 41% of these containing detectable transgenes in fins.

Growth of microinjected fish was quantitatively evaluated in three experiments. Average weight of microinjected fish was greater than that of controls of the same sex in four out of six groups. Signifi- cant growth enhancement (PC 0.05) was detected only for microinjected males in one experiment. Comparisons among molecular assays and individual fish growth in the founder generation indicated that the high degree of mosaicism prevented non-lethal indentification of all transgenic individuals and influences detection of growth enhancement.

Correspondence to: Dr. A.R. Kapuscinski, Fisheries and Wildlife, University of Minnesota, St. Paul, MN 55 108, USA. *Authors contributed equally; order was determined by a coin toss.

0044-8486/92/$05.00 0 1992 Elsevier Science Publishers B.V. All right- reserved.

254 ML. GROSS ET AL.

INTRODUCTION

The creation of growth-enhanced mice using gene transfer (Palmiter et al., 1982 ) has sparked interest in applying these techniques to improve traits in fish. While the classical method of improving traits is to develop a genetic selection program, the application of gene transfer has the potential to further enhance traits such as growth, disease resistance, and feed efficiency. Gene transfer experiments in fish have concentrated on growth enhancement using growth hormone genes (Guise et al., 1990; Kapuscinski and Hallerman, 1990). In some mammals containing growth hormone transgenes, increased growth hormone levels have caused deleterious perturbations in their endocrine balance resulting in serious side effects. For example, uncontrolled expression of growth hormone in pigs produced developmental problems that ranged from diabetes and infertility to lethality (Purse1 et al., 1989). Fish, however, may have the greatest promise for the action of transferred growth hormone genes since their rate of growth has been shown to be highly respon- sive to injections of crude, purified, or recombinant-derived growth hormone protein (Adelman, 1977; Agellon and Chen, 1986; Weatherly and Gill, 1987a,b).

Landings from the world’s fisheries are rapidly approaching the predicted maximum sustainable yield for natural waters. This trend has contributed to the increasing importance of aquaculture production (Stickney, 1990). Be- cause many aquaculture operations utilize wild or recently domesticated stocks, genetic improvements could potentially be financially beneficial. Sev- eral research groups have successfully introduced growth hormone genes into fish (Zhu et al., 1985; Chourrout et al., 1986; Dunham et al., 1987; Rokkones et al., 1989) with a few noting increased growth in the treated group (Zhu et al., 1986; Zhang et al., 1990). The latter reports, however, presented limited information about rearing conditions and statistical analyses, making it dif- ficult to quantitatively evaluate conclusions about the observed growth enhancement. In conjunction with molecular analyses, we therefore initiated growth studies under carefully controlled rearing conditions to quantitatively evaluate the performance of transgenic fish containing extra growth hormone genes.

MATERIALS AND METHODS

DNA constructs A fish expression vector (FV), F’&!/csGH, was constructed by inserting

the fragment containing chinook salmon (Oncorhynchus tshawytscha) growth hormone cDNA (Hew et al., 1989) into the EcoRI site of pHIN/Nco (Liu et al., 1990a). Thus, FV-2/csGH contains the proximal promoter and enhancer regulatory elements of the common carp (Cyprinus carpio) /Y-actin gene. To

NORTHERN PIKE MICROINJECTED WITH GROWTH HORMONE GENES 255

Hind Ill (partial)

Klenow dNTP’s

EcoR I linker

EcoR I

Hind Ill EcoR I

Clone into pHin/Nco

Figure Legend: E, EcoR I: H, Hind Ill; K, Kpn I; P, Pst I; S, Sst I; Sa, Sal I

carp n_ adin sequence @$f@ chinook salmon GH cDNA sequence

- vector sequences

Fig. 1. Construction of FV-2/csGH. Restriction sites are as indicated.

256 M.L. GROSS ET AL.

r Hinf I

Isolation of 330bp Hinf I fragment containing one copy of the LTR

Hinf I Hinf I

Clone into Hind Ill site of pUCll9

H ,/s= H ,,,:,::..<. .’ ‘.. :+:,>

0 pUCRSV

P

Pst I partial

Clone the complete copy of bGH cDNA (829bp) into pUCll9

r Hind Ill

Clone the isolated Hind Ill fragmant of LTR (343bp) into the Hind Ill site of pUCbGH

SP

Figure Legend:

B, SamH I: E, EcoR I: H, Hind Ill: Hf, Hinf I: K, Kpn I; P, Pst I; S, Sst I; Sa, Sal I; Sm, Sma I: Sp, Sph I; X, Xba I

- bGH 1-1 LTRof RSV - vector sequences

Fig. 2. Construction of pRSV/bGH. Restriction sites are as indicated.

NORTHERNPIKEMICROINJECTEDWITHGROWTHHORMONEGENES 257

insert the cDNA fragment into the EcoRI site of pHin/Nco, the clone that harbors the cDNA for c&H was partially digested with HindIII, filled-in with Klenow fragment DNA polymerase I, ligated to EcoRI linkers and digested with EcoRI endonuclease (Fig. 1).

To construct pRSV/bGH, the 0.83 Kb cDNA of bovine growth hormone was cloned into the PstI site of pUC 119 by partial digestion of pBH27 (pro- vided by Hank George, Molecular Genetics Inc., Edina, MN) with PstI to produce pUCbGH. The Rous Sarcoma virus (RSV) promoter contained within a 0.33 Kb Hi&I fragment (Yoon et al., 1990) was blunted with the Klenow fragment of polymerase I. Hind111 linkers were added and this fragment was cloned into the Hind111 site of pUCbGH to yield pRSV/bGH (Fig. 2).

Gamete collection, fertilization, and microinjection Northern pike (Esox lucius) gametes were collected from wild individuals

and transported at 4°C to the lab. Eggs from 2-3 females were mixed and fertilized with sperm from 2-3 males. A portion of the fertilized eggs was randomly allocated for controls. Twenty minutes after fertilization, a smoothly tapered borosilicate needle with an inner tip diameter of approximately 2 pm (Yoon et al., 1990) was used to microinject approximately 50 picoliters of linearized DNA solution into the germinal disc directly through the chorion and vitelline membranes. The plasmid pRSV/bGH was linearized with the restriction endonuclease KpnI, extracted with phenol/chloroform, precipi- tated in ethanol, and re-dissolved in 88 mM NaCl, 10 mMTris*HCl, pH 7.6 to a tinal concentration of 25 ng/& The plasmid FV-2/csGH was microin- jetted in the superhelical form at the same final concentration. Microinjec- tinn proceeded under a stereo microscope with the borosilicate needle at- tached to a Brinkman MM33 micromanipulator and the amount of plasmid solution injected was controlled by a Medical Systems Corporation pico-in- jector (Model PLI- 100).

DNA analysis The physical state of the transgene DNA was characterized by Southern

blot hybridization analysis, as described by Hallerman et al. ( 1990). DNA was isolated from individual B-day-old embryos or juvenile fish fin tissues by sodium dodecyl sulfate lysis (SDS) and proteinase K digestion. DNA (7 pg ) was digested with excess BamHI restriction endonuclease, electrophoresed through a 0.8% agarose gel, transferred onto nylon membranes by Southeru blotting and then probed for the presence of transgenic DNA. The entire plasmid, either pRSV/bGH or FV2/csGH, was labeled with 32P by the random priming method.

For two-dimensional gel electrophoresis, uncut and cellular DNA was electrophoresed first in 0.4% agarose and then at right angles in 1.2% agarose,


transferred to nylon membranes, and hybridized to labeled pRSV/bGH DNA (Manias et al., 1989). This method separates open circular and superhelical extra-chromosomal transgene DNA, both unit length and multimers, from linear cellular and integrated transgene DNA (Manias et al., 1989).

Growth hormone radioimmunoassay Blood serum samples collected from dorsal aortae of 2-month-old fish were

analyzed for the presence of bGH protein by radioimmunoassay. Because the pRSV/bGH construct included the signal peptide sequence, detection of bGH protein was expected. Radioiodinated bGH was prepared by adding four iodo- Beads (Pierce Chemical Co., Rockford, IL) to a vial containing 1.5 mCi’251 and 0.2 ml buffer (0.25 MNasP04 pH 7.5). After 5 min, 15 pg bovine growth hormone (USDA) solubilized in 0.01 MNaHC03 diluted with 20~1 of buffer was added and incubated for 15 min. The mixture was purified by column chromatography using Bio-gel P-30 (Bio-Rad) equilibrated and eluted with 0.1% bovine serum albumin (BSA)-barbitol buffer pH (Sigma) as three peaks. The middle peak was diluted in 1% BSA-0.01 M phosphate-buffered saline (PBS) pH 7.5. Anti-serum to ovine GH (NIAMDD-anti-oGH) was diluted 1: 5000 in 1% normal rabbit serum-O.05 M EDTA-PBS pH 7.0, an:: 0.1 ml dispensed into assay tubes containing 0.3 ml 1% egg albumin-PBS pH 7.0 and 0.3 ml test plasma. After 24 h, 0.1 ml tracer (20 000 cpm) was added, followed 24 h later by the addition of 0.1 ml precipitating antiserum (sheep anti-rabbit gamma globulin) and incubation for 48 h. Samples were centri- fuged at 15 OOOxg for 20 min and counted in a Beckman gamma counter. Assay sensitivity was approximately 1 ng/tube, equivalent to 1. P ng/ml (Wheaton et al., 1986).

Immunohistochemical detection of tissues expressing bGH Fin tissue collected from individual fish in the microinjected groups was

fixed, sectioned, and mounted on slides using standard clinical techniques. Slides were deparafinized by a 30 min room temperature dip in xylol, followed by two 5 min dips in lOO%, 90%, and 70% ethanol. Slides were washed three times in PBS pH 7.6 for 10 min, dipped in a 0.3% H202 in PBS for 30 min, and rewashed three times in PBS for 5 min. Blocking was by incubation in a 1 xdilution of normal sheep serum for 30 min at 37 “C, followed by a gentle wiping. Slides were incubated in polyclonal rabbit anti-bGH diluted 1: 100 in normal sheep serum overnight at 4”C, washed three times in PBS for 10 min, and incubated with goat anti-rabbit Ig diluted 1: 2000 in normal sheep serum for 30 min at room temperature. Slides were washed three times in PBS for 10 min, and incubated with the DAB (diaminobenzidine tetr;ihy- drochloride) substrate (5 mg DAB in 20 ml 0.03% H202 in PBS) for 10 min at room temperature in the dark. Slides were washed three times in PBS, counterstained in 0.15% methylene blue for 10 min, washed twice in tap water,


dehydrated by a series of dips in 70%, 95%, and 100% ethanol, dipped in xylol twice, and mounted with Permount.

Fish culture While optimum values for most culture variables have not been deter-

mined for northern pike, our culture methods were based on known require- ments of the species and empirical data ( Westers, 1986; Coolwater Fish Cul- ture Workshop Minutes, 1988, 1989, 1990). We set conservative values for maximum density index, minimum dissolved oxygen tension and maximum ammonia concentrations by using recommendations for rainbow trout (On- corhynchus mykiss), a species requiring relatively high water quality (Meade, 1985; Piper et al., 1982; Soderberg, 1982).

Incubation and hatching. Microinjected and control embryos were separately incubated in divided compartments of a flow-through vertical tray incubator. Embryos were incubated at 10°C until hatching (approximately 9 days). Im- mediately before hatching, embryos were placed in plastic pans in which the water was allowed to warm to room temperature, thereby reducing variability in fry size by synchronizing hatching of all embryos within an hour.

Fry rearing. Microinjected and control fry were separately reared in 2 10-l circular tanks at similar densities, ranging from 10 fry/l at hatching to less than 1 fry/l (or 15 g/l) at 2 months of age. Water temperature and flows were kept constant at 2 1 “C and one exchange per hour, respectively (Westers, 1986). After fry swim-up (approximately 3 days), automatic feeders delivered brine shrimp at 5-m!in intervals during 24 h of light. After 10 days, fry were converted to BioDiet pelleted feed (size # 2; BioProducts Inc. ), which was delivered by automatic Louden trough feeders (North Star Co.) at fre- quent intervals during light hours. Dual feeding of brine shrimp and BioDiet usually lasted 3 days.

Rearing after 2 months of age. Water temperatures, flow rates, and delivery of pelleted feed were as described for fry rearing. Densities ranged from 15 g/ 1 when average fish length was 12 cm to 95 g/l when average fish length was 30 cm. Photoperiod was 14 h light: 10 h dark and feed type Glencoe Trout Grower pellets (International Multifoods). When individual weights aver- aged 15 g, microinjected and control fish of an experiment were tagged intra- peritoneally with passive integrated transponders (Biosonics Corp.; Prentice et al., 1987) which permit individual identification. All microinjected and control individuals in a given growth experiment were then transferred to a common rearing tank.


TABLE 1

Percent hatch and numbers of fish retained for growth experiments

Expt. Construct Percent hatch

Injected Control

No. males analyzed

Injected Control

No. females analyzed

-

Injected Control

A pRSV/bGH 62. 58 22 7 24 9 B pRSV/bGH 63 78 38 17 98 26 C FV-2/csGH 66 75 61 7 67 24

Totals 64a 70” 121 31 189 59

‘Average value.

Growth analysis

A randomly chosen subset of the molecularly analyzed fish was compared to controls in three growth experiments (Table 1). Individual weights (to the nearest 0.1 g ) and lengths (to the nearest 1 .O mm) were recorded on a monthly basis. As fish gained. weight and space became limiting, numbers of fish in each experiment were reduced to maintain optimum rearing conditions. To cull fish, a computer-generated random number was assigned to each individual within the control or microinjected group in a common tank, the numbers were sorted and the appropriate numbers of fish were eliminated from the top of the list for the group.

Because females are usually larger than males and skewed sex ratios are known to exist in populations of northern pike (Hassler, 1969; Scott and Crossman, 1985 ), growth data were analyzed by sex. At 1 year of age, fish were sexed based on the shape of the urogenital opening (Demchenko, 1963; Casselman. 1974). Our ability to determine sex in this manner was verified by dissection of 100 fish. Weight differences between treatments of the same XX were analyzed by t-tests at approximately 6,9, and 12 months of age. F- statistics were calculated to test the equality of variances and determine whether pooled or separate variance t-tests applied (Zar, 1984).

XESULTS

By microinjection into the germinal disc of one-cell northern pike embryos, wi: obtained successful transfer and genomic integration of cDNA for either: bovine growth hormone linked to the Rous sarcoma virus long terminal repeat (pRSV/bGH; Fig. 2); or chinook salmon growth hormone linked to the carp pactin promoter (FV2/csGH; Fig. 1) . For some fish microinjected with pRSV/bGH, we detected expression of bGH in blood serum and growth enhancement under controlled growth trials. In earlier studies (Liu et al., 1990a),


we demonstrated expression of transgenes in fish under the transcriptional control of the carp @actin promoter.

Production of transgenic northern pike For the three growth experiments (Table 1 ), a total of 14 000 one-cell

northern pike embryos was microinjected. Hatching rates for both control (58-784/o) and microinjected (62-66%) groups were normal for artificial incubation of this species ( Westers, 1986). Of the surviving microinjected fish, 440 were randomly chosen for growth analyses (Table 2 ). In addition to these growth experiments, another 10 000 embryos were microinjected, of which 958 juvenile fish were molecularly analyzed (Table 2).

DNA analysis As shown in Fig. 3A, uncut genomic DNA from 6-day-old pWSV/bGH mi-

croinjected embryos appeared to be in two forms: linear copies ( 3.9 Kb) and high molecular weight DNA. The latter form, which represents the majority of the DNA, migrated as two separate bands: a band ( -23 Kb) representing large concatamers of the injected DNA, and a band larger than 40 Kb that migrated with cellular DNA and represents integrated transgenic DNA. In the lane for embryonic DNA digested with BarnHI several prominent bands exist: ( 1) the 3.9 Kb band represents linear copies of pRSV/bGH derived from digestion of both integrated and concatamer forms as well as the free linear copies present in the uncut lane; (2) the two faint high molecular weight bands most likely represent trimers and larger concatamers that have lost the BarnHI site during concatamerization; (3) the unexplainable 2.2 Kb band was found in all 6-day-old microiitjected embryos analyzed but not in control embryos (data not shown). The exact nature of the band is unknown at this point, but probably represents a common deletion product of the microinjected DNA.

TABLE 2

Numbers of presumptive juvenile transgenic fish molecularly analyzed and numbers in which transgene DNA or protein product (bGH ) was detected

Expt.B Fish analyzed DNA tin positive Serum bGH

A 52 B 260 C 128 Db 958

6(11%) 29 (11%) 9 (7%)

44 (5%)

nd 7 (3%)

nd 29 (3%)

Total 1398 88 (6%) 36 (3%)’

“DNA constructs used are described in Table 1. bThese fish were microinjected with pRSV/bGH but were not part of a growth experiment. ‘Represents 3% of total for experiments B and D ( 36112 18 ). nd = not done.

Juv 76 Juv 77

- 23

- 9.4

- 6.5

- 4.3

- 2.3 - 2.0

- 0.9

C Firs1 Dimension 0.4%

M.L. GROSS ET AL.

- 9.4

- 6.5

* - 2.3

- 2.0

Fig. 3. Southern blot hybridization analysis of DNA from selected northern pike transgenic for pRSV/bGH. (A) 10 pg of either BarnHI digested or uncleaved DNA from embryos was electrophoresed and transferred to nylon membranes, then hybridized to a [32P]-labelled probe made from a BumHI/Hin~II (0.83 Kb) fragment of pRSV/bGH containing the coding region of bGH. 20 pg of linearized pRSV/bGH cleaved with BumHI is included as a control. Lambda DNA cleaved with Hind111 was used for size markers (fragment sizes given in Kb). (B) 5 c(g of DNA from juvenile (Juv) northern pike flns’were cleaved with EcoRI, electrophoresed, transferred, then hybridized to a [“2P] labeled probe made from the entire pRSV/bGH plasmid. Expected band sizes corresponding to &oRI cleaved pRSV/bGH are 0.96 zi:d 2.9 Kb. Hind111 cleaved lambda phage DNA sewed as size markers (fragment sizes given in Kb). The bands at 2.5 and 2.1 Kb represent vector sequences from a co-microinjected pRSV/CAT plasmid (Hall- erman et al., 1990) included in some Juv fish. (Cl Two-dimensional gel electrophoresis analysis of uncut genomic DNA from Juvl7. Arrow indicates the expected location of episomal and concatamerized forms of DNA if they were present.

NORTHERN PIKE MlCROlNJECTED WITH GROWTH HORMONE GENES 263

Two-month-old juvenile fish were sufficiently large to allow removal of fin tissue without affecting survival. Therefore, to determine which of the surviving microinjected fish were transgenic, DNA was extracted from fin clips and screened for the presence of transgenic DNA by Southern blot hybridization. As shown in Table 2, 5-l 1% of microinjected fish in experiments A-D and 6% (88) of all surviving juvenile fish contained detectable levels of trans- genie DNA within their fin tissue. Unlike embryonic DNA, however, the transgenic DNA in juvenile fish appeared to be only in an integrated form. For example, in Fig. 3A, the pRSV/bGH transgenic sequences in the uncut Juv77 and Juv78 lanes only migrated as high molecular weight DNA. Fur- thermore, two-dimensional Southern blot hybridization analysis indicated that the transgenic DNA had probably been integrated into the genomes of juvenile fish (Fig. 3C). Transgenic DNA in the form of episomes or large super- coiled concatamers have been observed by other researchers; however, such forms would migrate above (see arrow Fig. 3C) the main bulk of the DNA. The possibility remains that some transgenic DNA is in a very long linear form, although such DNA generally does not have a long half-life in aevelop- ing organisms. Additionally, dominant bands from 70% orpRSV/bGH trans- genie fish DNA’s, when analyzed by Southern hybridization migrated at the expected band sizes: BamHI linearizes pRSV/bGH and produces a 3.9 Kb fragment; whereas EcoRI digestion gives bands of 0.96 and 2.9 Kb. For example, DNA of Juv78 digested with BamHI produced a dominant band at 3.9 Kb representing linear pRSV/bGH as well as many putative junction fragments (Fig. 3A). In Fig. 3B, all juvenile fish shown were digested with EcoRI and yielded the expected bands at 0.96 and 2.9 Kb. Putative junction fragments are also evident. However, other samples, including Juv77 (Fig.

3A) gave bands upon restriction digestion that were of unexpected sizes, sug-

gesting transgene rearrangement. Transgene copy number per fish ranged from less than one to hundreds of copies.

bGH assays

Because non-lethal removal of blood from 2-month-old fish was also possible, biood serum from these same fish was analyzed for the presence of circulating bovine growth hormone (Table 2 ) . Radioimmunoassays detected bGH hormone in the serum of 3% (36) of the tested fish. The majority of these positive fish had levels below 4 ng/ml of circulating bGH hormone within their serum, although nine fish had levels above 10 ng/ml with the highest being 37 ng/ml. Interestingly, 34 of the 36 fish had no detectable transgenic DNA within their fins, i r qplying a high degree of mosaicism within the transgenic fish.

Lack of any positive results with biood serum from control fish indicated that our bGH radioimmunoassay did not cross-react with native northern pike growth hormone. Additional evidence supporting this conclusion came


Fig. 4. Thin slice immunohistochemical analysis of northern pike fins employing a I: rabbit anti-bGW antibody. Panel A shows fin tissue from a control fish. Panel B shows from a fish microinjected with pRSV/bGH whose transgene expression is limited to entiated mesenchymal tissue, appearing as dark areas (especially prevalent within the cytes ). Abbreviations for tissue types are: (ch ) chondrocytes, (epi ) epithelial tissue, ray, and (urn) undifferentiated mesenchymal tissue.

)olyclonal i fin tissue undiffer- chondro- (ray) fin


TABLE 3

The extent of tissue DNA mosaicism in a random sample of northern pike microinjected with pRSV/ bGH

Fish no. Brain Gonad Kidney Liver Muscle Fin

1 14 16 19 20 21 24 26 27 29 31 36

+ + + + i-

+ + + + +

+ + + +

+ + +

f _-

+ +

+

J-

6/12 5112 TotaP 3/12 2/12 10/12

% of positive fish % of all analyzed fish

25 16 83 50 50 41

7 5 25 15 15 12

“12/40 fish were positive in one or more tissues, or 30°h overall.

from tests bwith a chinook salmon GH assay. This assay, which cross-reacted with northern pike blood, showed no cross-reactivity with purified bGH (Brian McKeown, Dept. of Biological Sciences, Simon Fraser University, Burnaby, B.C., personal communication).

Immunohistochemicul detection of bGH in fins

Two fish, with bGH in their serum but no detectable transgene in their fins, expressed bGH protein within their fins, as detected by a thin slice immunohistochemistry assay employing a polyclonal rabbit anti-bGH antibody (Fig. 4). Expression was restricted to a thin layer of mesodermal tissue within the fin, and was especially prevalent within the chondrocytes. The identical expression pattern was found in similar thin slices from several fish that were positive for both transgene DNA in fin tissue and for bGH in serum. At &is time, we do not know whether transgene DNA is present in a percentage of all the various types of fin cells, or only in those cells with observed expression of the hormone.

Detection of mosaicism Because transgenic fish appeared to be highly mosaic, we analyzed six tis-

sues (brain, gonad, kidney, liver, muscle, and fin) from 40 randomly selected

Exp. A


Oct1989 Jan1990 May1990

Ott 1989 Jan1990 May1990

220

8200 3 4 180

5 160

f 140

iii 120 3

i! 100

q 80

$ 60 $

40

20

0

Exp. C

Oct1989 Jan1990 May1990


200-

160-

160-

sr" 9 MO-

5 5 120-

w' loo- s

$ ao- d w

% 60-

40 -

20 -

O~.,.,.,.,.,.,.,.,.,.,.,.,.,.,~ 0 30 60 90 120 150 160 210 240 270 300 330 360 390 420

DAYS

Fig. 6. Average weight (g ) of males in Experiment A (pRSV/bGH) as a function of days since hatching. Asterisks denote t-test comparisons (*PC 0.05 and **PC 0.0 1) . Error bars represent twice the standard errors of means.

fish for the presence of transgenic DNA by Southern hybridization (Table 3 ) . Thirty percent of the fish from this sample contained detectable transgenic DNA in at least one or more tissues, with the kidney being the most prevalent (83% of positives). Only one fish (No. 27) was found to have DNA present within all examined tissues. Interestingly, transgene DNA was only detectable within fin tissues of 4 1% of the positive fish (or 12% of all examined fish), indicating our non-lethal screening procedure did not identify all transgenic individuals.

Growth analyses

Average weight of microinjected fish was greater than that of control fish of the same sex in four out of six groups (Fig. 5): females and males of experiment A, males of experiment B, and males of experiment C. At this point in

Fig. 5. Average weight (g), by sex, of microinjected and control fish for each growth experiment for three different time periods. DNA constructs used in each experiment are described in Table 1. The age of the fish, in days since hatching, were (for Expt. A-C): October 1989,167-182 days; January 1990,242-262 days; May 1990, 386-406 days. The P-values of t-test comparisons by sex for microinjected and control fish are superimposed on corresponding bars.


our ongoing study, however, microinjected males in experiment A are the only experimental group whose weights were significantly different (P< 0.05 ) from those of controls (Fig. 5). Length data (not shown) yielded the same trends as weight data in all experiments. Weights of microinjected and control males in experiment A steadily diverged over time (Fig. 6)) 280 a pattern also noted in females of experiment A and males of experiment C (data not shown). Similar patterns of divergence in growth have been observed in studies involving injections of bGH into fish ( Weatherly and Gill, 1987a; Schulte et al., 1989). Fish positive for either DNA within their tins or bGH expression within their blood were also compared as a group to controls by t-tests. Al- though they followed the same trends as seen for comparisons involving entire microinjected groups, weight differences from these group comparisons were insignificant (data not shown).

In every experiment the largest individuals of either sex were from the microinjected group, totalling 43 fish for the three experiments. Transgene DNA was detected in fin tissue of only three of these 43 fish. If the remainder of these fish are larger due to proper expression of transgenes, they must bear integrated and expressed transgenes in tissues undetectable by nonlethal methods.

DISCUSSION

Using standard techniques of microinjection into one-cell embryos, we obtained integration rates in juvenile fish (5-l 1% and 6% overall in fin tissue) comparable to those reported for common carp, C. curpio (5.5% in fin tissue; Zhang et al., 1990)) tilapia, Oreochromis nifoticus (6% in whole fish samples; Brem et al., 1988), and zebra&h, Brachydanio rerio (5% in fin tissue; Stuart et al., 1988 ). A higher integration rate ( 20% in whole fish samples) reported for juvenile channel catfish, Ictalurus punctatus (Dunham et al., 1987), was based on a sample size of 2 transgenic fish out of 10 microinjected embryos.

Our detection of circulating bGH in 3% of the analyzed microinjected fish indicated transgene expression but involved smali numbers of individuals ( I4 of each sex) within a growth experiment. Because this reduced statistical power, we did not conduct t-test comparisons of their growth to that of controls. We do not know if the minimum effective levels of bGH for growth enhancement are below or above the detection threshold of our assay, which was approximately 1 ng/ml. Individual levels of circulating bGH detected by our radioimmunoassays ranged from 2.0 ng/ml to 37.0 ng/ml. Reports of basal levels of endogenous growth hormone for fish include 7.2-58.4 ng/ml for chum salmon, Oncorhynchus keta (Bolton et al., 1986), lo-60 ng/ml for goldfish, Carassius auratus (Marchant and Peter, 1986), and lo-70 ng/ml for white sucker, Catostomus commersoni (Stacey et al., 1984). We also do not know how levels of hormone expression are developmentally regulated.


Given these unknowns, it is possible to miss detection of physiologically ac- tive levels of bGH in transgenic individuals showing growth enhancement.

Comparisons among data for integrated transgenes in fin DNA, serum bGI-I assays, and immunohistochemical detection of bGH in fin tissue indicated a high incidence of mosaic individuals in our founder generation of live trans- genie fish. This was confirmed by detection of transgene DNA in at least one of six examined tissues for 30% of 40 sacrificed fish. This is consistent with results of a previous study showing the distribution and expression of a marker transgene in goldfish (Hallerman et al., 1990). Stuart et al. ( 1988) also detected mosaicism in the founder generation of zebrafish based on the incidence of germ-line transmission and on the distribution of transgenes in eight tissues of one female. Patterns of germ-line transmission for the founder generation of common carp also indicated mosaicism (Zhang et al., 1990), although tissue comparisons of transgene integration and expression within individuals were not reported.

In this study we detected bGH expression in mesodermal tissue of fins and, in a previous report, we localized the prevalence of RSV-LTR-driven expression to mesodermal tissues in transgenic goldfish (Hallerman et al., 1990). Both reports parallel the incidence of tumorigenesis in such tissues following RSV virus infection into chickens and mice (Svet-Modalvsky, 1958; Pur- chase and Burmester, 1978 ). The molecular basis of this tissue tropism is not completely understood, but is presumed to be due to interactions between tissue specific transcription factors and the RSV-LTR. These same factors may therefore occur within fish, suggesting their high degree of conservation (Liu et al., 199Ob), as well as indicating the value of tra.tfsgenic fish as a model system for elucidating possible mechanisms in transcriptional control.

Our growth results represent a systematic evaluation of performance of founder generation transgenic fish under controlled rearing conditions. While the largest individuals in each experiment had been microinjected, significant growth enhancement (P~0.05) was observed only for microinjected males in one experiment (Figs. 5 and 6). Detection of statistically significant growth enhancement was influenced by three factors: presence of a mixture of genotypes in the founder generation of presumptive transgenic fish; variation in biological activity of the transgene product; and behavioral interactions.

Each microinjected group analyzed for growth performance was handled as one treatment for statistical analysis (F. Martin, Dept. of Applied Statistics, University of Minnesota, St. Paul, personal communication). However, each group actually consisted of several genotypes to which a given live individual could not be accurately assigned. Possible genotypes include: nontransgenic fish (probably > 50% based on results shown in Table 2 and 3); mosaics which lack proper transgene expression, thus performing like nontransgenics; mosaics which express the transgene in relevant tissues (many of which will be undetectable without sacrificing the fish); and completely transgenic individ-


uals which contain transgene DNA within all their cells and express it in a usable manner. The latter is the desired genotype but it probably occurs in only a small fraction of founder individuals using the current technique of microinjection at the one-cell stage of embryos.

Transgene expression leading to proper physiological activity is believed to require secretion of growth hormone into the bloodstream, whereby it is transported to target tissues and binds to cell surface receptors (Weatherly and Gill, 1987b). Therefore, if some founder generation individuals had expression of transgenes in non-target tissues without secretion into the bloodstream, they would not be expected to display growth enhancement. Lack of expression in some transgenic individuals could result from partial enzy- matic degradation of the construct prior to incorporation (Maclean et al., 1987) or from rearrangement of the construct. Our Southern hybridization data indicated rearrangement of integrated transgenes in some founder generation individuals.

At this time, we can only speculate about the impact of behavioral interactions within common rearing tanks on observed growth differences in our experiments. Aggressive behavior of a few individuals has been shown to sup- press growth of other fish in common tanks (Magnuson, 1962; Weatherly and Gill, 1987b). Additional experiments are needed to test specific hypotheses about the impact of behavioral interactions on growth differences between fish bearing growth promoting transgenes and controls.

As our fish reach sexual maturity, we are breeding them to produce populations of non-mosaic transgenic fish by germ-line transmission and are screening for progeny containing integrated sequences that are not rear- ranged. Elimination of mosaic genotypes will allow correct identification of transgenic individuals by non-lethal means, thus reducing ambiguities in assays for GH expression and assessment of growth performance. Current plans are to compare growth of transgenic progeny to controls in common and separate rearing tanks. This will permit partitioning of behavioral effects from direct physiological effects of growth hormone expressed by transgenes.

ACKNOWLEDGEMENTS

The authors would like to thank the Minnesota Department of Natural Re- sources for the provision of fish gametes, Dr. Choy Hew (Department of Clinical Biochemistry, University of Toronto) for supplying the chinook salmon growth hormone cDNA, and Dr. Ronald McGlennen, Dr. Brian McKeown, Dr. John Wheaton, Kris Carlson, Kurt LaBresh, Sung Joo Yoon, Zuoyan Zhu, Long He, Jay Maher, Mark Dalton and Mike Voss for their technical assistance and advice. The University of Minnesota Veterinary Patho- biology lab prepared slides of fin tissue sections. The anti-bGH antibodies were a gift from the National Hormone and Pituitary Program (University

NORTHERN PIKE MICROINJECTED WlTH GROWTH HORMONE GENES 271

of Maryland School of Medicine). Support for this work came in part from the Minnesota Agricultural Experiment Station, the Minnesota Sea Grant College Program supported by the NOAA Offke of Sea Grant, Department of Commerce under Grant Nos. NOAA-86AA-D-SG112, Project Nos. RA/3, RA/5, RA/6 and NOAA-90AA-SG149, Project Nos. RA/5, RA/6, RA/8, J. R. 267, the Legislative Commission on Minnesota Resources, The Blandin Foundation, and the Greater Minnesota Corporation. This is article 18 723 of the Minnesota Agricultural Experiment Station Scientific Journal Article Series.

REFERENCES

Adelman, I.R., 1977. Effect of bovine growth hormone on growth of carp (Cvprinus carp@ and the influences of temperature and photoperiod. J. Fish. Res. Board Can., 34: 509-S 15.

Agellon, L.B. and Chen, T.T., 1986. Rainbow trout growth hormone: molecular cloning of cDNA and expression in Escherichia coli. DNA, 5 (6): 463-47 1.

Bolton, J.P., Takahashi, A., Kawauchi, H., Kubota, J. and Hirano, T., 1986. Development and validation of a salmon growth hormone radioimmunassay. Gen. Comp. Endocrinol., 62: 230- 238.

Brem, G., Brenig, G., HGrstgen-Schwark, G. and Winnacker, E.L., 1988. Gene transfer in tilapia (Oreochromis niloticus). Aquaculture, 68: 209-2 19.

Casselman, J.M., 1974. External sex determination of northern pike, Esox Zucius Linnaeus. Trans. Am. Fish. Sot., 103( 2): 343-347.

Chourrout, D., Guyomard, R. and Houdebine, L., 1986. High efficiency gene transfer in rainbow trout (Salmo gairdneri Rich. ) by microinjection into egg cytoplasm. Aquaculture, 5 1: 143-150.

Coolwater Fish Culture Workshop Minutes, 1988. Aberdeen, South Dakota, 4-7 Jarmary 1988 (unpubl. ).

Coolwater Fish Culture Workshop Minutes, 1989. South Bend, Indiana, 10-l 1 January 1989 (unpubl.).

Coolwater Fish Culture Workshop Minutes, 1990. Bozeman, Montana, 8-10 January 1990 (unpubl.).

Demchenko, I.F., 1963. On sexual differences in the pike (Esox lucius). Vopr. Ikhtiol., 3: 190- 193 (in Russian).

Dunham, R., Eash, J., Askins, J. and Townes, T.M., 1987. Transfer of the metallothionein- human growth hormone fusion gene into channel catfish. Trans. Am. Fish. Sot., 116: 87-9 1.

Guise, K.S., Faras, A.J., Hackett, P.B. and Kapuscinski, A.R., 1990. Gene transfer in fish. In: N.I. First and F.S. Haseltine (Editors), Transgenic Animals, Proceedings of the Symposium on Transgenic Technology in Medicine and Agriculture. Butterworth-Heinemann, Boston, 358 pp.

Hallerman, E.M., Schneider, J.F., Gross, M., Liu, Z., Yoon, L., He, L., Hackett, P.B. and Ka- puscinski, A.R., 1990. Gene expression promoted by the RSV long terminal repeat element in transgenic goldfish. Anim. Biotech., 1 ( 1): 79-93.

Hassler, T.J., 1969. Biology of the northern pike in Oahe Reservoir, 1959 through 1965. U.S. Bureau of Sport Fisheries and Wildlife, Technical Paper, 29: l-l 3.

Hew, C.L., Trinh, KY., Du, S.J. and Song, S., 1989. Molecular cloning and expression of salmon pituitary hormones. Fish Physiol.. Biochem., 7: 375-380.


Kapuscinski, A.R. and Hallerman, E.M., 1990. Trans8enic fish and public policy. I. Anticipat- ing environmental impacts of transgenic fish. Fisheries, 15 ( 1): 2- 11.

Liu, z., MOW, B., Faras, A.J., Guise, K.S., Kapuscinski, A.R. and Hackett, P.B., 199Oa. Func- tional analysis of elements affecting expression of the P-actin gene of carp. Mol. Cell. Bid., lO( 7): 3432-3440.

Liu, Z., Moav, B., Faras, A.J., Guise, K.S., K .puscinski, A.R. and Hackett, P.B., 199Ob. Devel- opment of expression vectors for transgenic fish. Bio/Technology, 8: 1268- 1272.

Maclean, N., Penman, D. and Zhu, Z., 1987. Introduction of novel genes into fish. Bio/Tech- nology, 5: 257-26 1.

Magnuson, J.H., 1962. An analysis of aggressive behaviour, growth, and competition for food and space in medaka (O~~Z&T futipes) (Pisces, Cyprinodontidae). Can. J. Zool., 40: 313- 362.

Manias, D.A., Ostrow, R.S., McGlennen, R.C., Estensen, R.D. and Faras, A.J., 1989. Charac- terization of integrated human papillomavirus type I 1 DNA in primary and metastatic tu- mors from a renal transplant recipient. Cancer Res., 49: 25 14-25 19.

Marchant, T.A. and Peter. R.E., 1986. Seasonal variations in body growth rates and circulating levels of growth hormone in the goldfish, Curussius aurutus. J. Exp. Zool., 237: 23 l-239.

Meade, J.W., 1985. Allowable ammonia for fish culture. Prog. Fish-Cult., 47 ( 3 ): 135- 145. Palmiter, R.D., Brinster, R.L., Manner, R.E., Trumbauer, M.E., Rosenfeld, M.C., Birnberg,

N.C. and Evans, R.M., 1982. Dramatic growth of mice that develop from eggs microinjected with metallothionein growth hormone fusion genes. Nature, 300: 6 1 l-6 15.

Piper, R.G., McElwain, LB., Orme, L.E., McCror en, J.P., Fowler, L.G. and Leonard, J.R., 1982. Fish Hatchery Management. U.S. Dept. of the Interior, Fish and Wildlife Service, Washing- ton, DC, 5 17 pp.

Prentice, E.F., Flagg, T.A. and McCutcheon, S., 1987. A study to determine the biological fea- sibility of a new fish tagging system, 1986-87. Report to Bonneville Power Admin., Project 83-3 19, Seattle, I 13 pp.

Purchase, H.G. and Burmester, B.R., 1978. Leukosis/sarcoma group. In: M.S. Hofstad, B.W. Calnek, C.F. Helmboldt, W.M. Reid and H.W. Yoder (Editors), Diseases of Poultry. Iowa State Univ. Press, Ames, 949 pp.

Pursell, V.G., Pinkest, C.A., Miller, K.F., Bolt, D.J., Campell, R.G., Palmiter, R.D., Brinster, R.L. and Hammer, R.E., 1989. Genetic engineering of livestock. Science, 244: I28 I-I 288.

Rokkones, E., Alestrom, P., Skervold, H. and Gautvik, K.M., 1989. Microinjection and expression of a mouse metallothionein human growth hormone fusion gene in fertilized salmonid eggs. J. Comp. Physiol. B., 158: 75 1-758.

Schulte, P.M., Down, N.E., Donaldson, E.M. and Souza, L.M., 1989. Experimental administra- tion of recombinant bovine growth hormone to juvenile rainbow trout (Salmo guirdneri) by injection or by immersion. Aquaculture, 76: 145- 156.

Scott, W.B. and Crossman, E.J., 1985. Freshwater Fishes of Canada. Fisheries Research Board of Canada Bulletin 184, Ottawa, 966 pp.

Soderberg, R.W., 1982. Aeration of water supplies for fish culture in flowing waters. Prog. Fish- Cult., 44: 89-93.

Stacey, N.E., MacKenzie, D.S., Marchant, T.A., Kyle, A.L. and Peter, R.E., 1984. Endocrine changes during natural spawning in the white sucker, Cutostomus cornmersoni. I. Gonado- tropin, growth hormone, and thyroid hormones. Gen. Comp. Endocrinol., 56: 333-348.

Stickney, R., 1990. A global overview of aquaculture production. Food Rev. Int., 6: 299-3 ; 5.

Stuart, G.W., McMurray, J.V. and Westerfield, M., 1988. Replication, integration, and stable germ line transmission of foreign sequences injected into early zebrafish embryos. Develop- ment, 103: 403-4 12.

Svet-ModaIvsky, G.J., 1958. Sarcoma in albino rats treated during the embryonic stage with Rous virus. Nature, 182: 1452- 1453.

NORTHERN PKE MICROINJECTED WITH GROWTH HORMONE GENES 273

Weatherly, A.H. and Gill, H.S., 1987a. Growth increases produced by bovine growth hormone in grass pickerel, Esox americanus vermiculafus ( LeSueur j, and the underlying dynamics of muscle fiber growth. Aquaculture, 65: 55-66.

Weatherly, A.H. and Gill, H.S., 1987b. The Biology of Fish Growth. Academic Press, London, 443 pp.

Westers, H., 1986. Northern pike and muskellunge. In: R. Stickney (Editor), Culture of Non- salmonid Freshwater Fishes. CRC Press, Boca Raton, FL, 20 1 pp.

Wheaton, J.E., Al-Raheem, S.N., Massri, Y.G. and Marcek, J.M., 1986. Twenty-four hour growth hormone profiles in Angus steers. J. Anim. Sci., 62: 1267- 1272.

Yoon, S.J., Hallerman, E.M., Gross, M.L., Liu, Z., Schneider, J.F., Faras, A.J., Hackett, P.B., Kapuscinski, A.R. and Guise, KS., 1990. Transfer of the gene for neomycin resistance into goldfish, Carassius auratus. Aquaculture, 85: 2 l-23.

Zar, J.H., 1984. Biostatistical Analysis. Prentice-Hall, Englewood Cliffs, NJ, 7 18 pp. Zhang, P., Hayat, M., Joyce, C., Gonzalez-Villaseiior, L.I., Lin, C.M., Dunham, R.A., Chen,

T.T. and Powers, D.A., 1990. Gene transfer, expression and inheritance of pRSV-rainbow trout-GH cDNA in the common carp, Cyprinus carpio (Linnaeus). Mol. Reprod. Dev., 25: 3-13.,

Zhu, Z., Li, G., He, L. and Chen, S., 1985. Novel gene transfer into the fertilized eggs ofgoldfish (Carassius auratus L. 1758). Z. Angew. Ichthyol., 1: 31-34.

Zhu, Z., Xu, K., Li, G., Xie, Y. and He, L., 1986. Biological effects of human growth hormone gene microinjected into the fertilized eggs of loach Msgurnus anguillicaudafus. Kexue Tong- bao Acad. Sin., 3 1 ( 14): 988-990.

molecular analysis and growth evaluation of northern pike

Documents